Estimating bradykinesia severity in Parkinson's disease by analysing gait through a waist-worn sensor

Abstract

Bradykinesia is a cardinal symptom of Parkinson's disease (PD) and describes the slowness of movement revealed in patients. Current PD therapies are based on dopamine replacement, and given that bradykinesia is the symptom that best correlates with the dopaminergic deficiency, the knowledge of its fluctuations may be useful in the diagnosis, treatment and better understanding of the disease progression. This paper evaluates a machine learning method that analyses the signals provided by a triaxial accelerometer placed on the waist of PD patients in order to automatically assess bradykinetic gait unobtrusively. This method employs Support Vector Machines to determine those parts of the signals corresponding to gait. The frequency content of strides is then used to determine bradykinetic walking bouts and to estimate bradykinesia severity based on an epsilon-Support Vector Regression model. The method is validated in 12 PD patients, which leads to two main conclusions. Firstly, the frequency content of the strides allows for the dichotomic detection of bradykinesia with an accuracy higher than 90%. This process requires the use of a patient-dependant threshold that is estimated based on a leave-one-patient-out regression model. Secondly, bradykinesia severity measured through UPDRS scores is approximated by means of a regression model with errors below 10%. Although the method has to be further validated in more patients, results obtained suggest that the presented approach can be successfully used to rate bradykinesia in the daily life of PD patients unobtrusively.

A Waist-Worn Inertial Measurement Unit for Long-Term Monitoring of Parkinson’s Disease Patients

“…The extracted features are, firstly, temporal characteristics such as mean, standard deviation, range, signal magnitude area, signal correlations, distribution analysis like skewness, kurtosis; and secondly frequency features are used, such as energy and spectral density in specific bands. From these features the bradykinetic gait [41,64] and dyskinesia indexes are extracted [61], as well as if the person is walking based on an SVM model [41,64,66], and the FoG detection output [60]. The calculated index for dyskinesia and bradykinesia, however, might be dismissed given that the person might not suffer neither dyskinesia nor bradykinesia in a certain moment of the day.…”

“…Regarding the assessment of bradykinesia, the symptom evaluation is widely described in [64,66]. The feature extraction consists of extracting energy spectral density of a concrete band for detecting gait.…”

“…The fluidity computation score is based on frequency features and is compared to a threshold trained by means of a leave one out process [41,64]. …”

A Kinematic Sensor and Algorithm to Detect Motor Fluctuations in Parkinson Disease: Validation Study Under Real Conditions of Use (Preprint)

“…The SVM had been previously trained with labeled signals from 20 different PD patients, who did not participate in the On and Off state monitoring for data collection in this study. Walk detection accuracy was higher than 90% [23]. 2.…”

Gold nanostructures electrodeposited on graphene oxide-modified indium tin oxide glass as a surface-enhanced Raman scattering-active substrate for ultrasensitive detection of dopamine neurotransmitter

“…[5][6][7] Several diseases of the nervous system are associated with the dysfunctions or the deficiency of the DA system, such as schizophrenia and Huntington's and Parkinson's disease. [8][9][10] Patients with parkinsonism have lost more than 80% of the dopamine-producing cells in the substantia nigra. DA detection is helpful for the early diagnosis of the neurologic diseases mentioned above and for evaluating drug efficiency.…”